{"id":"https://openalex.org/W2246078565","doi":"https://doi.org/10.1145/2814895.2814931","title":"An automatic speech detection architecture for social robot oral interaction","display_name":"An automatic speech detection architecture for social robot oral interaction","publication_year":2015,"publication_date":"2015-10-07","ids":{"openalex":"https://openalex.org/W2246078565","doi":"https://doi.org/10.1145/2814895.2814931","mag":"2246078565"},"language":"en","primary_location":{"id":"doi:10.1145/2814895.2814931","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2814895.2814931","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Audio Mostly 2015 on Interaction With Sound","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5032140560","display_name":"Emmanouil Tsardoulias","orcid":"https://orcid.org/0000-0001-9034-4832"},"institutions":[{"id":"https://openalex.org/I4210134249","display_name":"Centre for Research and Technology Hellas","ror":"https://ror.org/03bndpq63","country_code":"GR","type":"facility","lineage":["https://openalex.org/I4210134249"]}],"countries":["GR"],"is_corresponding":true,"raw_author_name":"E. G. Tsardoulias","raw_affiliation_strings":["Centre of Research & Technology - Hellas, 6th km Xarilaou - Thermi, 57001, Thessaloniki, Greece"],"affiliations":[{"raw_affiliation_string":"Centre of Research & Technology - Hellas, 6th km Xarilaou - Thermi, 57001, Thessaloniki, Greece","institution_ids":["https://openalex.org/I4210134249"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047313464","display_name":"Andreas L. Symeonidis","orcid":"https://orcid.org/0000-0003-0235-6046"},"institutions":[{"id":"https://openalex.org/I21370196","display_name":"Aristotle University of Thessaloniki","ror":"https://ror.org/02j61yw88","country_code":"GR","type":"education","lineage":["https://openalex.org/I21370196"]},{"id":"https://openalex.org/I4210134249","display_name":"Centre for Research and Technology Hellas","ror":"https://ror.org/03bndpq63","country_code":"GR","type":"facility","lineage":["https://openalex.org/I4210134249"]}],"countries":["GR"],"is_corresponding":false,"raw_author_name":"A. L. Symeonidis","raw_affiliation_strings":["Centre of Research & Technology - Hellas, 6th km Xarilaou - Thermi, 57001 / Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, 54124, Thessaloniki, Greece"],"affiliations":[{"raw_affiliation_string":"Centre of Research & Technology - Hellas, 6th km Xarilaou - Thermi, 57001 / Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, 54124, Thessaloniki, Greece","institution_ids":["https://openalex.org/I4210134249","https://openalex.org/I21370196"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5014906615","display_name":"Pericles A. Mitkas","orcid":"https://orcid.org/0000-0003-2090-4202"},"institutions":[{"id":"https://openalex.org/I21370196","display_name":"Aristotle University of Thessaloniki","ror":"https://ror.org/02j61yw88","country_code":"GR","type":"education","lineage":["https://openalex.org/I21370196"]}],"countries":["GR"],"is_corresponding":false,"raw_author_name":"P. A. Mitkas","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, 54124, Thessaloniki, Greece"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, 54124, Thessaloniki, Greece","institution_ids":["https://openalex.org/I21370196"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5032140560"],"corresponding_institution_ids":["https://openalex.org/I4210134249"],"apc_list":null,"apc_paid":null,"fwci":1.2943,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.86339119,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9993000030517578,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.9991999864578247,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7854606509208679},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.609965980052948},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5795133709907532},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5647017359733582},{"id":"https://openalex.org/keywords/robotics","display_name":"Robotics","score":0.5639547109603882},{"id":"https://openalex.org/keywords/sphinx","display_name":"Sphinx","score":0.5639033317565918},{"id":"https://openalex.org/keywords/voice-activity-detection","display_name":"Voice activity detection","score":0.5336056351661682},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.488591730594635},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.47354716062545776},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4674699306488037},{"id":"https://openalex.org/keywords/humanoid-robot","display_name":"Humanoid robot","score":0.43320295214653015},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.37894871830940247},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.12564441561698914},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08427852392196655}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7854606509208679},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.609965980052948},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5795133709907532},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5647017359733582},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.5639547109603882},{"id":"https://openalex.org/C2777299493","wikidata":"https://www.wikidata.org/wiki/Q151480","display_name":"Sphinx","level":2,"score":0.5639033317565918},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.5336056351661682},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.488591730594635},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.47354716062545776},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4674699306488037},{"id":"https://openalex.org/C60692881","wikidata":"https://www.wikidata.org/wiki/Q584529","display_name":"Humanoid robot","level":3,"score":0.43320295214653015},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.37894871830940247},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.12564441561698914},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08427852392196655},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/2814895.2814931","is_oa":false,"landing_page_url":"https://doi.org/10.1145/2814895.2814931","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Audio Mostly 2015 on Interaction With Sound","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.7799999713897705,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W5726830","https://openalex.org/W72347498","https://openalex.org/W1491238342","https://openalex.org/W1524333225","https://openalex.org/W1637669343","https://openalex.org/W2017133406","https://openalex.org/W2024733869","https://openalex.org/W2050365646","https://openalex.org/W2062609023","https://openalex.org/W2075139806","https://openalex.org/W2096284813","https://openalex.org/W2099739380","https://openalex.org/W2122028591","https://openalex.org/W2128653836","https://openalex.org/W2136657283","https://openalex.org/W2146842127","https://openalex.org/W2147550046","https://openalex.org/W2157725521","https://openalex.org/W2165874967","https://openalex.org/W2182193419","https://openalex.org/W2741763241","https://openalex.org/W4254373586","https://openalex.org/W4365800070"],"related_works":["https://openalex.org/W642007152","https://openalex.org/W2401827384","https://openalex.org/W2355290951","https://openalex.org/W4304187160","https://openalex.org/W2052688117","https://openalex.org/W2552102772","https://openalex.org/W4294771049","https://openalex.org/W1523214805","https://openalex.org/W2168417340","https://openalex.org/W4229451372"],"abstract_inverted_index":{"Social":[0],"robotics":[1,8],"have":[2,32],"become":[3],"a":[4,17,29,38,66,81,119,150,154],"trend":[5],"in":[6,16,40,96,133],"contemporary":[7],"research,":[9],"since":[10,63],"they":[11],"can":[12],"be":[13],"successfully":[14],"used":[15],"wide":[18],"range":[19],"of":[20,23,69,73,111,162],"applications.":[21],"One":[22],"the":[24,34,58,109,112,137,159],"most":[25],"fundamental":[26],"communication":[27],"skills":[28],"robot":[30],"must":[31],"is":[33,51,118,144],"oral":[35],"interaction":[36],"with":[37],"human,":[39],"order":[41],"to":[42,121,158],"provide":[43],"feedback":[44],"or":[45,128],"accept":[46],"commands.":[47],"And,":[48],"although":[49],"text-to-speech":[50],"an":[52],"almost":[53],"solved":[54],"problem,":[55],"this":[56],"isn't":[57],"case":[59],"for":[60,124],"speech":[61,105,141],"detection,":[62],"it":[64],"includes":[65],"large":[67],"number":[68],"different":[70,104],"conditions,":[71],"many":[72],"which":[74,125],"are":[75,79,168],"literally":[76],"unpredictable.":[77],"There":[78],"quite":[80],"few":[82],"well":[83],"established":[84],"ASR":[85],"(Automatic":[86],"Speech":[87],"Recognition)":[88],"tools,":[89],"however":[90],"without":[91],"providing":[92],"efficient":[93],"results,":[94],"especially":[95],"less":[97],"popular":[98],"languages.":[99],"The":[100,116,140],"current":[101],"paper":[102],"investigates":[103],"detection":[106,142],"strategies":[107],"via":[108],"utilization":[110],"Sphinx-4":[113],"open-source":[114],"library.":[115],"first":[117],"way":[120],"incorporate":[122],"languages":[123],"no":[126],"acoustic":[127],"language":[129],"model":[130,143],"exists":[131],"(Greek":[132],"our":[134],"case),":[135],"following":[136],"grapheme-to-phoneme":[138],"concept.":[139],"evaluated":[145],"using":[146],"audio":[147],"captured":[148],"from":[149],"NAO":[151],"v4":[152],"robot,":[153],"difficult":[155],"task":[156],"due":[157],"high":[160],"levels":[161],"included":[163],"noise,":[164],"thus":[165],"denoising":[166],"techniques":[167],"investigated":[169],"as":[170],"well.":[171]},"counts_by_year":[{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":1},{"year":2018,"cited_by_count":1},{"year":2017,"cited_by_count":1},{"year":2015,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}